PHARMACOKINETICS BY FLOW CYTOMETRY: RECOMMENDATIONS FOR DEVELOPMENT AND VALIDATION OF FLOW CYTOMETRIC METHOD FOR PHARMACOKINETIC STUDIES - Flow Cytometry in Drug Discovery and Development - page 227

temperature, wash, sample acquisition cell number, and data analysis. Tolerance
limits (assay robustness) should also be established for critical steps.
Selection of an appropriate cell number (per staining) would allow a drug standard
curve with a broad dynamic range. Lack of prozone effect (or hook effect) where lower
binding signal is seen at high drug concentrations is also important to investigate. The
use of DOE is encouraged as it allows efficient assessment of multiple parameters,
especially when these parameters interact with each other. Details on the use of DOE
are described by Anderson and Whitcomb [2].

11.2.7.2 Assay Format Assays can be developed using either a tube or a
microtiter plate format. The use of tubes results in a relatively low throughput,
but allows a more thorough washing by using higher volume wash buffer. The use of
96-well plate for staining is more efficient, but due to the limited volume of wash
buffer that can be added to each well, more than one wash may be needed. Evaluation
of incubation time for each step and establishment of tolerance limits as part of assay
robustness are important to define assay conditions. Each sample should be tested in
duplicate or triplicate wells and results averaged. The average MFI (median) is
considered a single data point with a CV

15% being acceptable.

11.2.7.3 Staining Condition Staining can be done in PBS containing BSA and
sodium azide at 4 C. Room-temperature staining should be performed in a
temperature-controlled (22-26 C). incubator Incubation times should be

determined to obtain maximal binding signal without increase in background.

11.2.7.4 Specificity and Selectivity Specificity is the ability of an analytical
method to exclusively detect the target analyte and display complete discrimina-
tion from physiologically/chemically similar compounds that may be present.
Specificity considerations for the secondary detection antibody are previously
described. It is good practice to evaluate binding to drug across the standard curve
range with drug prepared in buffer versus in sample matrix (serum, EDTA plasma,
heparin plasma, etc.). The use of the appropriate sample matrix (pooled serum or
plasma that matches the study sample) to prepare reference standard drug and QC
samples is recommended. If there is a sample matrix effect on the assay, interpolating
data of study sample from a standard curve using drug in buffer may result in over- or
underdetection.
Selectivity is examined to assess spike and recovery. Drugs at H, M, and L were
spiked into sample matrix (serum or plasma) from multiple donors and results
compared to samples spiked in buffer. Recovery of 80-120% is considered accep-
table. Recoveries outside this range should be used with caution and warrant further
interpretation of results. An interference study can be done to examine potential
soluble interfering factors that could be present in patient's samples and influence
results. For this type of study, drugs at H, M, and L levels were added to sample matrix
in the presence of potential soluble interference factors (each at H, M, or L levels).
This study can include sample composition interference, such as hemolysis or
lipemia, or more specific interference such as soluble binding proteins or receptors.